An Introduction to Word Embeddings.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# An Introduction to Word Embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One of the breakthroughs of neural networks in Natural Language Processing is the usage of word embeddings. Rather than using the words themselves as features, neural network methods typically take as input dense, relatively low-dimensional vectors that model the meaning and usage of a word. Word embeddings were first popularized through the [Word2Vec](https://arxiv.org/abs/1301.3781) model, developed by Thomas Mikolov and colleagues at Google. Since then, scores of alternative approaches have been developed, such as [GloVe](https://nlp.stanford.edu/projects/glove/) and [FastText](https://fasttext.cc/) embeddings. In this notebook, we'll explore word embeddings with the original Word2Vec approach, as implemented in the [Gensim](https://radimrehurek.com/gensim/) library. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training word embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training word embeddings with Gensim couldn't be easier. The only thing we need is a corpus of sentences in the language under investigation. Wikipedia is a good choice for training generic embeddings. For our experiments, we're going to use 5,000,000 sentences from Dutch Wikipedia, which we've trained and lowercased in advance. This means we can feed lists of sentence tokens to Word2Vec by reading the lines in our Wikipedia file and splitting them on spaces."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "class SentenceCorpus(object):\n",
    "\n",
    "    def __init__(self, filename):\n",
    "        self.filename = filename\n",
    "\n",
    "    def __iter__(self):\n",
    "        with open(self.filename, \"r\") as i:\n",
    "            for line in i:\n",
    "                tokens = line.strip().split()\n",
    "                yield tokens\n",
    "                \n",
    "                \n",
    "WIKI_FILE = os.path.join(\"../data\", \"nlwiki_20170620_tok_small.txt\")\n",
    "sentences = SentenceCorpus(WIKI_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When we train our word embeddings, gensim allows us to set a number of parameters. The most important of these are `min_count`, `window`, `size` and `sg`:\n",
    "\n",
    "- `min_count` is the minimum frequency of the words in our corpus. For infrequent words, we just don't have enough information to train reliable word embeddings. It therefore makes sense to set this minimum frequency to at least 10. In these experiments, we'll set it to 100 to limit the size of our model even more.\n",
    "- `window` is number of words to the left and to the right that make up the context that word2vec will take into account.\n",
    "- `size` is the dimensionality of the word vectors. This is generally between 100 and 1000. You often have to make a trade-off: embeddings with a higher dimensionality are able to model more information, but also need more data to train.\n",
    "- `sg`: there are two algorithms to train word2vec: skip-gram and CBOW. Skip-gram tries to predict the context on the basis of the target word; CBOW tries to find the target on the basis of the context. By default, Gensim uses CBOW (`sg=0`).\n",
    "\n",
    "We'll investigate the impact of some of these parameters later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gensim\n",
    "\n",
    "model = gensim.models.Word2Vec(sentences, min_count=100, window=5, size=100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Using word embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at the model. The word embeddings are on its `wv` attribute, and we can access them by the using the token as key. For example, here is the embedding for Dutch *koning* (king), with the requested 100 dimensions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 2.507609  , -0.30699807,  0.47751993, -0.4576649 ,  2.1689956 ,\n",
       "        1.3362927 ,  1.0770283 ,  0.8022279 ,  0.60743684, -2.064534  ,\n",
       "       -1.4264114 ,  2.1700997 ,  0.02347353,  0.2881159 , -0.23293515,\n",
       "       -1.0216573 ,  0.5559268 , -1.2184445 ,  0.11585803,  0.78457665,\n",
       "       -4.1839867 , -0.18230717,  5.4035335 ,  1.2598838 ,  2.201747  ,\n",
       "       -1.2833642 ,  0.6694852 , -1.1582412 ,  0.6086354 , -0.16085055,\n",
       "       -2.6774156 , -3.1560042 ,  2.82665   ,  2.14405   ,  2.4224594 ,\n",
       "       -0.8019148 ,  0.22807464, -1.6112757 , -0.16821289, -2.8398442 ,\n",
       "       -0.36864278, -3.005337  , -0.32036808, -2.5830822 ,  0.43935654,\n",
       "        1.5883299 , -0.5146567 , -2.0608008 , -0.13783525,  1.168791  ,\n",
       "       -0.6834477 ,  2.0023494 , -0.24786559,  0.47408956,  1.2373208 ,\n",
       "        0.49597648,  2.8643968 ,  1.1659172 ,  0.87470126, -0.61956084,\n",
       "        0.37980655, -0.629845  ,  0.7331885 , -2.8164258 ,  0.95166934,\n",
       "        0.4734735 ,  2.948517  ,  1.9941856 , -1.3659931 , -0.4929753 ,\n",
       "       -1.0052358 ,  2.5825775 , -1.2698338 ,  0.4433992 ,  1.9003807 ,\n",
       "        1.7921144 ,  0.530539  , -2.8560529 , -2.9380252 , -1.913015  ,\n",
       "       -2.0588653 , -3.1619277 , -1.1980247 , -2.3895388 ,  4.366712  ,\n",
       "       -1.882515  ,  0.08873676, -1.7349355 , -4.7019796 ,  3.4950883 ,\n",
       "       -3.0734918 ,  0.447328  ,  0.62204325,  1.1164886 , -0.5575447 ,\n",
       "        1.3246256 ,  2.3032079 ,  0.12082383,  1.1538653 , -0.092497  ],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv[\"koning\"]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also easily find the similarity between two words. Similarity is measured as the cosine between the two word embeddings, and ranges between -1 and +1. The higher the cosine, the more similar two words are. As expected, the figures below show that *koning* (king) is closer to *koningin* (queen) than to *koffie* (coffee)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7425808\n",
      "0.0236511\n"
     ]
    }
   ],
   "source": [
    "print(model.wv.similarity(\"koning\", \"koningin\"))\n",
    "print(model.wv.similarity(\"koning\", \"koffie\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In a similar vein, we can find the words that are most similar to a target word. The words with the most similar embedding to *koning* are all similar titles (such as *keizer* (emperor) and *hertog* (duke)) or are semantically related to royalty (such as *troon* (throne))."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('keizer', 0.8405033349990845),\n",
       " ('vorst', 0.8119637370109558),\n",
       " ('kroonprins', 0.8083297610282898),\n",
       " ('stadhouder', 0.7592759728431702),\n",
       " ('hofmeier', 0.7552233934402466),\n",
       " ('hertog', 0.7534068822860718),\n",
       " ('landvoogd', 0.7511402368545532),\n",
       " ('troonopvolger', 0.7502525448799133),\n",
       " ('troonpretendent', 0.7460724115371704),\n",
       " ('koningin', 0.7425808906555176)]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.similar_by_word(\"koning\", topn=10)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Interestingly, we can look for words that are similar to a set of words and dissimilar to another set of words at the same time. This allows us to look for analogies of the type *king (koning) is to man (man) like ... is to woman (vrouw)*. Although the most similar word is not the correct answer (which would be queen), notice how female titles, such as *echtgenote* (wife), *keizerin* (empress) and *koningin* (queen) are now present in the top 10 most similar words. This wasn't the case above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('kroonprins', 0.7478970885276794),\n",
       " ('koningin', 0.7189857959747314),\n",
       " ('keizerin', 0.6947569847106934),\n",
       " ('troonopvolger', 0.6943620443344116),\n",
       " ('echtgenote', 0.6906800270080566),\n",
       " ('groothertog', 0.6882891654968262),\n",
       " ('onderkoning', 0.6744648218154907),\n",
       " ('vorstin', 0.6732239127159119),\n",
       " ('hertog', 0.6702529191970825),\n",
       " ('isabella', 0.6694217920303345)]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar(positive=['vrouw', 'koning'], negative=[\"man\"], topn=10)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly, we can also zoom in on one of the meanings of ambiguous words. For example, like in English, *muis* (mouse) in Dutch can refer to two things: an animal and a computer mouse. If we look at the 10 nearest neighbours to *muis*, most of them are animals, such as *papegaai* (parrot). This suggests the animal meaning is much more frequent on Wikipedia than the other one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('schotel', 0.7593557834625244),\n",
       " ('kat', 0.7543159127235413),\n",
       " ('geit', 0.7396884560585022),\n",
       " ('papegaai', 0.7295396327972412),\n",
       " ('slang', 0.7294548749923706),\n",
       " ('eend', 0.721156656742096),\n",
       " ('hond', 0.7205180525779724),\n",
       " ('aap', 0.7156603336334229),\n",
       " ('ezel', 0.7011622786521912),\n",
       " ('dief', 0.6958762407302856)]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar(positive=[\"muis\"], topn=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, if we specify we're looking for words that are similar to *muis* (mouse), but dissimilar to *dier* (animal), suddenly the computer meaning takes over. We now find similar devices in the top ten nearest neighbours, such as *afstandsbediening* (remote control)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('afstandsbediening', 0.48941680788993835),\n",
       " ('laptop', 0.4737693965435028),\n",
       " ('controller', 0.45984235405921936),\n",
       " ('driver', 0.4594345688819885),\n",
       " ('tetra', 0.4513355493545532),\n",
       " ('usb', 0.4467567205429077),\n",
       " ('r8', 0.4410499036312103),\n",
       " ('client', 0.43251797556877136),\n",
       " ('krijg', 0.4308721125125885),\n",
       " ('sql', 0.4281649887561798)]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar(positive=[\"muis\"], negative=[\"dier\"], topn=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we can present the word2vec model with a list of words and ask it to identify the odd one out. It then uses the word embeddings to identify the word that is least similar to the other ones. For example, in the list *auto fiets bus koffie* (car, bike, bus, coffee), it correctly identifies *koffie* as the odd one out. In the list *koffie auto thee melk* (coffee, car, tea, milk), it correctly singles out *auto*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "koffie\n",
      "auto\n"
     ]
    }
   ],
   "source": [
    "print(model.wv.doesnt_match(\"auto fiets bus koffie\".split()))\n",
    "print(model.wv.doesnt_match(\"koffie auto thee melk\".split()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now visualize some of our embeddings. To plot embeddings with a dimensionality of 100 or more, we first need to map them to a dimensionality of 2. We do this with the popular [t-SNE](https://lvdmaaten.github.io/tsne/) method. T-SNE, short for t-distributed Stochastic Neighbor Embedding, helps us visualize high-dimensional data by mapping similar data to nearby points and dissimilar data to distance points in the low-dimensional space.\n",
    "\n",
    "T-SNE is present in [Scikit-learn](https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html). To run it, we just have to specify the number of dimensions we'd like to map the data to (`n_components`), and the similarity metric that t-SNE should use to compute the similarity between two data points (`metric`). We're going to map to 2 dimensions and use the cosine as our similarity metric. Additionally, we use PCA as an initialization method to remove some noise and speed up computation. The [Scikit-learn user guide](https://scikit-learn.org/stable/modules/manifold.html#t-sne) contains some additional tips for optimizing performance. \n",
    "\n",
    "Plotting all the embeddings in our vector space would result in a very crowded figure where the labels are hardly legible. Therefore we'll focus on a subset of embeddings by selecting the 200 most similar words to a target word. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "target_word = \"belgië\"\n",
    "selected_words = [w[0] for w in model.wv.most_similar(positive=[target_word], topn=200)]\n",
    "embeddings = [model.wv[w] for w in selected_words]\n",
    "\n",
    "mapped_embeddings = TSNE(n_components=2, metric='cosine', init='pca').fit_transform(embeddings)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we take *België* (Belgium) as our target word, the figure shows some interesting patterns. Notice how all countries can be found at one side of the picture, with the European countries closer together. In other parts of the picture, Belgian provinces cluster closely together, as well as some provinces. In all, the distances between the words give us considerable confidence in the quality of our embeddings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x2160 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(30,30))\n",
    "x = mapped_embeddings[:,0]\n",
    "y = mapped_embeddings[:,1]\n",
    "plt.scatter(x, y)\n",
    "\n",
    "for i, txt in enumerate(selected_words):\n",
    "    plt.annotate(txt, (x[i], y[i]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploring hyperparameters"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We mentioned above there are a number of parameters when can set when training our embeddings. Let's investigate the impact some of these have on the result. Quantifying the quality of embeddings is a hard task. There exist quite a few data sets for [evaluating the quality of English embeddings](https://github.com/kudkudak/word-embeddings-benchmarks), but this is not the case for other languages. Moreover, it's unclear what information good embeddings should capture. Should they model syntactic information as well as semantic knowledge? Should they capture semantic similarity, or merely topical relatedness? Often, the answer depends on the end task you want to use the embeddings for. \n",
    "\n",
    "Here we'll use a simple method for evaluating our embeddings. We'll count how many times two nearest neighbours in the vector space have the same part of speech. After all, if our embeddings are any good, we expect a noun to have another noun as nearest neighbour, and the same for verbs, adjectives, and so on.\n",
    "\n",
    "First we'll use [spaCy](https://spacy.io/) to determine the part of speech of all the words in our vocabulary. Note that our evaluation metric does rely on the quality of spaCy's part-of-speech tagging, which may not be very accurate for low-frequency words out of context. Nevertheless, we'll assume it's good enough for our purposes. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dd6eeeab2fef44a4af08bbe9cc024f6f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'NOUN'"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import spacy\n",
    "from tqdm import tqdm_notebook as tqdm\n",
    "\n",
    "nl = spacy.load(\"nl\")\n",
    "\n",
    "word2pos = {}\n",
    "for word in tqdm(model.wv.vocab):\n",
    "    word2pos[word] = nl(word)[0].pos_\n",
    "\n",
    "word2pos[\"koning\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we write a simple method that takes a model and looks up the nearest neighbour to every word in its vocabulary. It then returns the number of times this nearest neighbour has the same part of speech: a percentage we'll call the accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4cb1b929500446d2ac19ad16b43cbce4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.6395543481243654"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "def evaluate(model, word2pos):\n",
    "    same = 0\n",
    "    for word in tqdm(model.wv.vocab):\n",
    "        most_similar = model.wv.similar_by_word(word, topn=1)[0][0]\n",
    "        if word2pos[most_similar] == word2pos[word]:\n",
    "            same += 1\n",
    "    return same/len(model.wv.vocab)\n",
    "       \n",
    "evaluate(model, word2pos)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we vary some of the settings we introduced above. In particular we're interested in the influence of embedding size (the dimensionality of the trained embeddings), and the size of the context window. We vary the embedding size between 100, 200 and 300,and the context window between 2, 5 and 10. This means we'll train 9 models in total, which obviously takes a bit of time. Go grab a coffee. Or two. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Size: 100 Window: 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "098602d889084366a90948e744456951",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 100 Window: 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9e319d7c2e984a069d852bd3533b943a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 100 Window: 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "75d6e5ab83c84aa084b3122aeb438f15",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 200 Window: 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3be811b0393c40e69a501445bf71e2d0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 200 Window: 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0530f2344d6d401c8c84bb9887b64275",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 200 Window: 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "81fc8cedba5746deb2a4829357d5fc4e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 300 Window: 2\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "246f3350d1e0450dbc658ae5ed2613ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 300 Window: 5\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1c69eb90aa3c4605a76fba7fa9118fa1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Size: 300 Window: 10\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e41d56a0824d4dd79dc25f00abd16083",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, max=36441), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>100</th>\n",
       "      <th>200</th>\n",
       "      <th>300</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.64164</td>\n",
       "      <td>0.643616</td>\n",
       "      <td>0.639609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.636618</td>\n",
       "      <td>0.63681</td>\n",
       "      <td>0.637716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.633188</td>\n",
       "      <td>0.636261</td>\n",
       "      <td>0.633929</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         100       200       300\n",
       "2    0.64164  0.643616  0.639609\n",
       "5   0.636618   0.63681  0.637716\n",
       "10  0.633188  0.636261  0.633929"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sizes = [100, 200, 300]\n",
    "windows = [2,5,10]\n",
    "\n",
    "df = pd.DataFrame(index=windows, columns=sizes)\n",
    "\n",
    "for size in sizes:\n",
    "    for window in windows:\n",
    "        print(\"Size:\", size, \"Window:\", window)\n",
    "        model = gensim.models.Word2Vec(sentences, min_count=100, window=window, size=size)\n",
    "        acc = evaluate(model, word2pos)\n",
    "        df[size][window] = acc\n",
    "        \n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the accuracies of all models are very similar, the results do show some interesting patterns. \n",
    "\n",
    "First, it looks like smaller contexts work better than larger ones. This is logical, as our evaluation metric is a syntactic one: the closest context words contain much more useful information about the part of speech of a word than those further away. \n",
    "\n",
    "Second, 200-dimensional context embeddings seem to work best. This may sound counter-intuitive, as higher-dimensional embeddings are able to capture more information. Still, such higher-dimensional embeddings also require more data. For the relatively small corpus we're working with, 200-dimensional embeddings strik the perfect balance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f941d7ffa90>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Clustering embeddings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we're going to cluster our embeddings. This will be useful for our experiments with Named Entity Recognition later, where embedding clusters allow Conditional Random Fields to model semantic information. We'll use agglomerative clustering, a bottom-up clustering method that iteratively takes together the two most similar clusters (or embeddings) in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import AgglomerativeClustering\n",
    "from sklearn.preprocessing import normalize\n",
    "\n",
    "vocab = list(model.wv.vocab)\n",
    "vectors = [model.wv[w] for w in vocab]\n",
    "vectors_norm = normalize(vectors)\n",
    "\n",
    "clusterer = AgglomerativeClustering(n_clusters=500)\n",
    "clusters = clusterer.fit_predict(vectors_norm)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's inspect some of the clusters. By focusing some of the clusters that contain the names of countries, we can see how these clusters can be useful for Named Entity Recognition. When a CRF encounters a word it has never seen, it can rely on the fact that it is present in a country cluster, to label it as a location. The same goes for other entity types. We'll save the clusters to a file so we can use them later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "cluster_dictionary = {}\n",
    "for cluster, word in zip(clusters, vocab): \n",
    "    if cluster not in cluster_dictionary:\n",
    "        cluster_dictionary[cluster] = []\n",
    "    cluster_dictionary[cluster].append(word)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['duitsland', 'frankrijk', 'oostenrijk', 'italië', 'zwitserland', 'slovenië', 'liechtenstein', 'joegoslavië', 'spanje', 'polen', 'turkije', 'luxemburg', 'macedonië', 'sovjet-unie', 'rusland', 'ussr', 'griekenland', 'portugal', 'tsjecho-slowakije', 'tsjechië', 'hongarije', 'tsjechoslowakije', 'roemenië', 'bulgarije', 'denemarken', 'noorwegen', 'zweden', 'finland', 'oekraïne', 'oost-duitsland', 'bondsrepubliek', 'servië', 'west-duitsland', 'armenië', 'bosnië', 'cyprus', 'wit-rusland', 'litouwen', 'letland', 'albanië', 'slowakije', 'ddr', 'georgië', 'oost-berlijn', 'west-berlijn', 'estland', 'montenegro', 'kroatië', 'herzegovina', 'vojvodina']\n"
     ]
    }
   ],
   "source": [
    "for x in cluster_dictionary:\n",
    "    if \"italië\" in cluster_dictionary[x]:\n",
    "        print(cluster_dictionary[x])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"data/clusters_nl.tsv\", \"w\") as o:\n",
    "    for c in cluster_dictionary:\n",
    "        for w in cluster_dictionary[c]:\n",
    "            o.write(f\"{w}\\t{c}\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Word embeddings are one of the most exciting trends on Natural Language Processing since the 2000s. They allow us to model the meaning and usage of a word, and discover words that behave similarly. This is crucial for the generalization capacity of many machine learning models. Moving from raw strings to embeddings allows them to generalize across words that have a similar meaning, and discover patterns that had previously escaped them."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}